winlab summary sept 7, 2011 - winlab.rutgers.edu€¦ · winlab summary sept 7, 2011 wireless...
TRANSCRIPT
1
WINLAB Summary
Sept 7, 2011
Wireless Information Network Laboratory (WINLAB)
Rutgers, The State University of New Jersey
www.winlab.rutgers.edu
Contact: Professor D. Raychaudhuri, Director
WINLAB
WINLAB
Introduction: Mission & Resources WINLAB founded in 1989 as a collaborative industry-university research
center with specialized focus on wireless networking
Mission is to advance both research and education in the area of wireless technology (… a topic of fast growing importance across the entire information technology field!)
Research scope includes information theory, radio technology, wireless networks, mobile computing and pervasive systems
Participation in several major federal research initiatives in the wireless and networking fields - cognitive radio/spectrum, future Internet architecture (FIA), GENI
Unique Rutgers resource with local, national and international recognition and impact
WINLAB resources in brief:
~25 faculty/staff, most from the ECE and CS departments at Rutgers
~40-50 grad students (80% PhD, 20% MS) – ~50 PhD’s graduated since 2005; ~20 UG internships
~$5M/yr research funding (80% federal, 20% industry); ~15 corporate sponsors from all over the world
~25,000 sq-ft facility, mostly at the Rt 1 Technology Center building (see photo)
Unique experimental capabilities including ORBIT testbed (see photo) and WiNC2R cognitive radio
ORBIT Radio Grid Testbed WINLAB Tech Center Facility
WINLAB 3
WINLAB Summary: Industry Sponsors
*
*Research Partners
InPoint
Semandex
Igolgi
*
*
*
Panasonic
US Army CECOM
*
WINLAB
WINLAB Summary: People
Dipankar
Raychaudhuri Roy Yates Narayan Mandayam Chris Rose Wade Trappe Predrag Spasojevic Yanyong Zhang Marco Gruteser Ivan Seskar
Athina Petropulu Larry Greenstein Dick Frenkiel Rich Howard Richard Martin
Yicheng Lu
Melissa Gelfman Noreen DeCarlo Janice
Campanella Elaine Connors
Khanh Le
Shridatt
Sugrim ~40-PhD & MS
Students as of 2011
(see www.winlab.rutgers.edu for photos)
Kiran Nagarja Sam Nelson
Ilya Chigivev Jaskaran Singh
Hui Xiong Zoran Miljanic Jun Li Michael Littman Mor Naaman
WINLAB 5
WINLAB Summary: Research Vision Radio everywhere from ~1B wireless devices in 2005 to
~10B in 2010 100B in 2020! Fundamental capacity and scale limits
Overcoming spectrum scarcity
New technology foundation – cognitive radios
Wireless – Internet convergence into a single global network as mobile terminals replace PC’s Architectural implications of mobility, disconnection, location, …
Wireless “network-of-networks” with heterogeneous radios, multi-hop, etc.
Clean-slate protocol architecture centered around mobility & context
From basic voice/data communications pervasive computing Wireless as the glue for integrating the Internet with the physical world
Importance of geographic location as a key attribute
Security and privacy
Various application domains – transportation, healthcare, security, industrial automation, …
WINLAB
WINLAB Summary: Research Scope
Static Spectrum
Assignment
Dynamic Spectrum
Assignment
~10x eficiency
Single User
MIMO/OFDM
Next-Gen
Gigabit PHY
Static MAC
Protocols
Flexible &
Adaptive MAC
IP Routing +
Cellular Mobility
Mobility-Centric
Internet Arch
Mobile web
services
Content- and context-aware pervasive
services
Spectrum sensing, NC-OFDM,
Spectrum server, cognitive algorithms,
Coordination protocols, ..
Network MIMO, network coding,
interference alignment, 60 Ghz,
Cooperative relay, cross-layer,
beam switching, software MAC,..
Storage-aware routing, global name
resolution, location, vehicular nets,
privacy/security, ad hoc/DTN routing, …
Content- and context-aware protocols, M2M
Programmable networks, cloud services, …
WINLAB
WINLAB Summary: Research Scope Dynamic Spectrum Assignment (DSA)
Spectrum policy models and coexistence algorithms
Spectrum sensing, databases and protocols for coordination
“Cognitive” Software-Defined Radio (SDR) Core technology for next-generation wireless systems
Flexible, high-performance architecture (WiNC2R, GENI SDR)
Next-generation wireless & the future Internet Clean-slate mobility-centric Internet architecture (MobilityFirst)
New protocol concepts: Storage routing, global name service, ..
Mobile network privacy and security aspects
Pervasive computing protocols & applications Active RFID for object tracking and location determination
Geographic protocols for vehicular and sensor networks
Future wireless networking testbeds ORBIT radio grid and outdoor GENI WiMAX (4G cellular)
Cognitive radio networking protocols and testbed (CogNet)
Bluetooth-WiFi Spectrogram
WiNC2R SDR Platform
Extracting Secret Key from Radio Signal
Vehicular Radio Node
GENI WiMAX Base Station
8
Research Project
Samples
WINLAB
channel 25 availability
Fixed
TV white space in NJ using 10X10 grid
Number of channels available in each square area
Number of channels vs. number of areas
Portable
White Space: Analysis of Available Spectrum
WINLAB
White Space: Secondary Co-
Existence Methods
WS Mobile
Access Protocol
WS AP
w/ backhaul
Secondary System A Secondary System B
freq
Secondary A Spectrum
Secondary B
Spectrum
Secondary co-existence an important requirement for white space bands
Various schemes possible depending on system model
Completely autonomous, using performance feedback only
Common coordination channel or Internet-based spectrum service
Common Coordination Channel (optional)
Internet
Spectrum Server (optional)
Control
information
NSF and sponsor
Funded projects
(P. Spasojevic,
I. Seskar, D. Raychaudhuri)
WINLAB
Rechargeable Networks: Optimal Retransmission
Policies Jing Lei, Zhuo Chen, Roy Yates
Picture Courtesy of S. Roundy, UC Berkeley
Energy replenishment rate is stochastic and environment-constrained
Rechargeable battery has a long life but its energy should not be abused
Message transmission carries different rewards
Our goal is to maximize the average reward rate by selective transmission .
WINLAB
Mobiles have intermittent network connections During disconnection, network caches in-transit personal content
At reconnection, content retrieved from nearest cache
Users specify acceptable caching prices
Routers advertise caching prices for LRU cache service
Prices set to support an expected lifetime for cached objects
Reduced transit times
Increased cache hit ratios
Mobile Content Delivery: Pricing for Caching Personal Content
WINLAB 13
GeoMac achieves lower delay and more reliable transmissions than ad hoc routing protocols
Unreliable wireless channel: severe fading due to shadowing
due to static and mobile obstructions
GeoMac opportunistically uses neighboring vehicles to forward
messages. Forwarders selected based on
geographic heuristics
V2V Networks: GeoMAC for Reliable Broadcast Messaging
WINLAB
Optical V2V: Visual MIMO
Explores MIMO free space optical communications with camera receivers
First analytical results show higher capacity than traditional FSO in mobile setting
Transmitter Array Receiver Array
Co-channel Interference free!
NSF-funded project
(Marco Gruteser, N. Mandayam & K. Dana)
WINLAB
Vehicular Applications: Drive-By
Sensing (ParkNet) Goal: Low-cost collection of road-side
parking availability
Uses GPS and ultrasonic rangefinders (42 kHz, 20 samples/s, 15cm resolution, up to 6.5m distance)
Preliminary results show 90% accuracy with threshold detection algorithm
NSF-funded project
(Marco Gruteser & collaborators)
WINLAB 16
Pervasive Systems: Tracking with
Roll Call RFID
Cancer Clinic 80x100 ft
Paper chart Pipsqueak 2.0 tag
1 second beacon interval
Median accuracy 12 ft.
Fixed costs of $2.50
sq/ft.
WINLAB
0.00
1.00
2.00
3.00
4.00
5.00
6.00
7.00
8.00
0 20 40 60 80 100 120 140 160
Mo
bili
ty S
core
Seconds
Tag 8e Mobile
Tag 77 Stationary
Tag 3B Mobile
Threshold
Pervasive Systems: Mobility Detection
Using Active Tags
WINLAB
Security & Privacy: User Controlled
Wireless Privacy
Always-on wireless use through smartphones, automobiles allows ubiquitous tracking Privacy metrics, criteria for
system evaluation?
Providing notice and accountability – allow users to detect when tracking takes place
Strong privacy techniques to thwart localization and tracking
NSF-funded projects
(Marco Gruteser & collaborators)
WINLAB
MobilityFirst: Robust & Trustworthy Mobility-
Centric Architecture for Future Internet
Base Station
Wireless Router
AP
Core Network
(flat label routing)
Router
Global Name Resolution Service
Control & Management Plane
Computing Blade
Buffer Storage
Forwarding Engine
MobilityFirst
Router with
Integrated
Computing & Storage
Hop-by-Hop
Transport
GDTN Routing
Name <-> Net address mapping
Data block Data
Plane
MobilityFirst key protocol
features: Separation of naming & addressing
Fast global naming service
Storage-aware (GDTN) routing
Hop-by-hop (segmented) transport
Self-certifying public key names
Support for content/context/location
Programmable computing layer
Separate network mgmt plane
New components, very
distinct from IP, intended to
achieve key mobile Internet
design goals
Multi-institutional NSF project
Led by WINLAB
WINLAB
MobilityFirst: Name-Address Separation
Separation of names (ID) from
network addresses (NA)
Globally unique name (GUID)
for network attached objects User name, device ID, content, context,
AS name, and so on
Multiple domain-specific naming
services
Global Name Resolution Service
for GUID NA mappings
Hybrid GUID/NA approach Both name/address headers in PDU
“Fast path” when NA is available
GUID resolution, late binding option
Globally Unique Flat Identifier (GUID)
John’s _laptop_1
Sue’s_mobile_2
Server_1234
Sensor@XYZ
Media File_ABC
Host
Naming
Service
Network
Sensor
Naming
Service
Content
Naming
Service
Global Name Resolution Service
Network address
Net1.local_ID
Net2.local_ID
Context
Naming
Service
Taxis in NB
WINLAB
MobilityFirst: Packet Headers and
Forwarding with Hybrid GIUD/NetAddr Primary design option under consideration is a hybrid scheme with support for
both name (GID) and topological address (NA) routing
NA header used for “fast” path, with fallback to GUID resolution where needed
Facilitates generalized multicast and late binding services
GUID/Service
Header
Data Object
GID-Address Mapping
Routing Table
GID NA
12345.. xxyy, xxzz
GUID/Servce
Header
Data Object
NA Header
Dest NA Path
xxyy Net1, net2, ..
Flat GID Routing
(slow path)
Topological Address Routing
(fast path)
Name-GID Mapping
Name GID
server@winlab Net123.localxyz GID/Service
Header
Data File
NA Header
Data Object
PDU with name PDU with name
and address
GUID/Public Key
Hash
SID
(Service
Identifier)
GUID Header Components
*Note: MobilityFirst PDU contains the entire data object
ranging from short message to large media file ~ GB
with hop-by-hop transport and storage (explained later)
WINLAB
MobilityFirst: GUID/Address Routing
Scenarios – Dual Homing The combination of GUID and network address helps to support new mobility
related services including multi-homing, anycast, DTN, context, location …
Dual-homing scenario below allows for multiple NA:PA’s per name
Data Plane
GUID
DATA
Send data file to “Alice’s laptop”
Net 1
Net 7
Current network addresses provided by GNRS;
NA1:PA22 ; NA7:PA13
GUID NA1:PA22; NA7:PA13
DATA
GUID NA1:PA22; NA7:PA13
DATA
Router bifurcates PDU to NA1 & NA7
(no GUID resolution needed)
Dual-homed
mobile device
GUID NetAddr= NA7.PA13
DATA
GUID NetAddr= NA1.PA22
DATA
Alice’s laptop
GUID = xxx
WINLAB
Experimental Systems: Platforms & Testbeds
Degree of Realism
Scale
Math models
Network Science
Opnet or ns Simulator
WINLAB ORBIT Radio Grid Emulator
Open Cellular
Campus Testbeds
GENI Core
Network
CO-WINLAB CR Platform USRP2
USRP/GNU Radio
SDR Sandbox in ORBIT
NSF Funded ORBIT & GENI
Projects at WINLAB
WINLAB
Experimental Systems: ORBIT Outdoor
Testbed Infrastructure
•Distributed across three campuses in NJ (and campus in Australia connected over
L2 tunnel)
•Mixture of production and experimental traffic
“Experiments”:
•Dynamic allocation of resources
•Development of “virtual node”: uses devices that belong to different
testbeds
Virtualized WiMAX
WINLAB
Experimental Systems: Virtual WiMAX
Networks
Design Goals: Multiple independent virtual
networks (VNs), each with specified % of BS capacity
Inter-slice fairness & isolation
For GENI experiments, each VN should be qualitatively equivalent to a dedicated BS
Each VN (slice) should support multiple clients
Intra-slice fairness
Multiple traffic types
Ph
ysi
cal
80
2.1
6e
BS
10
%
30
%
20
%
Slice1 Slice2 Slice3
WINLAB
Experimental Systems: Virtual WiMAX
Network Implementation for GENI
• VN traffic shaping (VNTS) on external GENI controller
• Maintains fairness & isolation between slices
• Uses SNMP status feedback (MCS, rate,..) from BS
No Shaping
VNTS